Proposal Title
Collaborative testing: evidence of learning in a controlled in-class study of undergraduate students, or "more learning while testing, less time only testing"
Session Type
Poster
Room
PAB Atrium
Start Date
9-7-2013 5:30 PM
Keywords
collaborative testing, two-stage exam, assessment, test
Primary Threads
Evaluation of Learning
Abstract
In collaborative two-stage exams students complete a test as individuals and then immediately complete the same, or very similar, test in groups. We compared two-stage collaborative testing to individual testing to determine which format has a greater effect on student learning in an undergraduate Earth and Ocean Science course. A cross-over design allowed students to participate in both the control (individual) and treatment (collaborative) conditions. In both the individual and collaborative conditions students completed the same set of questions twice, which controlled for any potential performance gain caused by more frequent testing. Learning was measured as the change in students’ individual performance on questions given in the individual stage and after the midterm, calculated as percent change and normalized change. When students were tested in groups they showed significantly greater improvement on subsequent individual testing then when tested only as individuals. There was no significant difference in the amount of improvement experienced by ‘upper’, ‘middle’ or ‘lower’ achieving students as categorized by their first-stage midterm score. Our results demonstrate that students of all achievement levels can learn from each other while simultaneously being assessed, and we believe collaborative testing is a powerful tool that can be applied to any undergraduate science classroom.
Collaborative testing: evidence of learning in a controlled in-class study of undergraduate students, or "more learning while testing, less time only testing"
PAB Atrium
In collaborative two-stage exams students complete a test as individuals and then immediately complete the same, or very similar, test in groups. We compared two-stage collaborative testing to individual testing to determine which format has a greater effect on student learning in an undergraduate Earth and Ocean Science course. A cross-over design allowed students to participate in both the control (individual) and treatment (collaborative) conditions. In both the individual and collaborative conditions students completed the same set of questions twice, which controlled for any potential performance gain caused by more frequent testing. Learning was measured as the change in students’ individual performance on questions given in the individual stage and after the midterm, calculated as percent change and normalized change. When students were tested in groups they showed significantly greater improvement on subsequent individual testing then when tested only as individuals. There was no significant difference in the amount of improvement experienced by ‘upper’, ‘middle’ or ‘lower’ achieving students as categorized by their first-stage midterm score. Our results demonstrate that students of all achievement levels can learn from each other while simultaneously being assessed, and we believe collaborative testing is a powerful tool that can be applied to any undergraduate science classroom.